Research Interests
*wordle from text from paper titles and abstracts
My primary research area is mathematical modeling of biological systems, particularly applications in ecology and infectious diseases. Most of my current research focuses the general areas of:
- Inference methods (primarily Bayesian) for mechanistic models of biological systems
- The ecology of infectious diseases
- The effects of spacial heterogeneity, dispersal, and environmental
stochasticity
on population dynamics and population persistence - Optimal life history strategies
Below I describe some of my current and previous research more specifically. I am currently studying chytridomitosis in frogs (Rana muscosa), as well as the application of Dynamic Energy Budget theory to Daphnia. For both of these projects, a major component is developing methods for parameter inference. I previously studied adaptive intervention strategies for infectious diseases and the implications of various dispersal strategies on persistence of money spider populations. My dissertation work focused on cholera, and modeling bacterial aging and biofilm formation.
Papers and technical reports on the work described here can be found on my Publications page.
Population effects of Chytridiomycosis in
Rana muscosa and R. sierrae (mountain and Sierra yellow-legged
frogs)
Chytridiomycosis,
a fungal disease of amphibians, has been implicated in
widespread population declines and extinctions in amphibian populations
worldwide. Efforts to understand the pathogenesis and global spread of the
disease have been wide ranging -- from intensive observation of field
populations and laboratory experiments to mathematical and statistical modeling.
However, many aspects of the disease, such as infection rates (both between and
within frogs) and reproductive rates of the fungus, are very difficult to
measure directly, especially in natural populations. Yet, these factors are
crucial for understanding why some populations are heavily affected by the
disease while others are not, as well as for designing effective control
strategies. Thus we want to be able to build models that can reproduce observed
patterns, and use these to attempt to extract information about these processes
that are difficult to measure directly. A mechanistic IBM that includes these
factors has been explored by Briggs et al.
(2010) to
further our understanding
of the dynamics of the disease as the fungus grows on individual frogs and is
transmitted between them. This model includes many components that we know are
important for understanding the observed infection patterns. However, few
statistical approaches to inference for IBMs are available. I am working on
developing statistical methodologies for performing parameter inference for the
particular IBM for this system, as well as more generally.
For more information on the research being conducted in the lab, see the
Briggs Lab Research page.
Articles/Information in the Popular Press
- New York Times article on the threatened Sierra yellow-legged frog
- National Geographic Article on vanishing amphibian populations.
- PBS Nature Special on the decline of frog populations.
Dynamic Energy Budget Theory
Dynamic Energy Budget (DEB) theory, at a basic level, seeks to
describe how organisms uptake and use energy for physiological
processes, such as growth, maintenance, and reproduction, using
formal mechanistic models. These models take
into account chemical and physical constraints (such as conservation of
mass and energy, and
homeostasis), as well as biological detail. These properties make the
theory broadly
applicable, and allow us to better understand how energetic
considerations shape patterns observed in nature.
I am developing tools to perform Bayesian inference for DEB
models that include dynamic food availability (through forcing, feedback, or both).
A frequent simplifying assumption in the analysis of DEB models is that the food
environment experienced by an individual is constant. However, it is well known
that this condition rarely holds. However the impact of making
such assumptions on our understanding of important physiological parameters is poorly
understood. We explore this problem using a combination of simulated and real data on
Daphnia, a water flea.
We presented initial results on this work at the 2nd International
Symposium on
Dynamic Energy Budget Theory in Lisbon in April 2011. If you would like copies of
the slides, etc., please e-mail me.
More information on DEB in general is available HERE.
Adaptive Management of Epidemiological Interventions
Typically, there are three major goals of mathematical models of infectious
diseases: to understand why we see the patterns we do; to be able to predict
future outbreaks; and to be able to intervene in an epidemic in an effective
way. Thus, we want to be able to design models of disease proliferation that
capture important dynamics, can be parameterized with data, and can used to
develop more effective control strategies. Sometimes, such as during the
spread of an emerging disease or a new strain of an existing disease, key
epidemiological parameters may be unknown, making the formulation of
effective control strategies difficult. My collaborators and I developed a
framework of Bayesian methods for on-line estimation of epidemiological
parameters and for adaptive management of a disease outbreak. This framework
allows us to update our knowledge of what is going on as a disease spreads,
and use the new information to modify intervention strategies as an epidemic
progresses. As part of this work, we also developed an implementation of the
framework in R, called "amei", which is available on CRAN.
Two papers (one on the methodology and one on the software package) are
linked from my publications page, as is the
software package.
Spider Dispersal and Population Persistance
In my previous position I studied Linyphiid, or money, spiders. These spiders are an important part of agricultural ecosystems; they eat aphids and other pests, and are a food source for birds. However, populations of these spiders appear to be decreasing in the United Kingdom. It is hypothesized that a major part of the decline is due to intensive agricultural practices, such as increased pesticide use.
The goal of this research is to explore how spider behaviors - particularly "ballooning", a long distance dispersal strategy - influence the persistence of populations in heterogeneous, stochastic environments. When ballooning, a small spider floats on air currents using a single strand of web. Long distance dispersal complicates the effects of local dynamics on the population, so it is important to know how far the spiders can travel. However, this is impossible to observe directly. I have developed a quasi-IBM that combines stochastic local population dynamics with a data driven dispersal model. With this model I have been able to determine how dispersal influences population persistence in the face of field level catastrophes (such as pesticide application).
See my publications page for a link to my paper on this work.
Cholera and Vibrio cholerae
In my dissertation, I studied the dynamics of cholera on three scales: cholera in a human population coupled to a reservoir of the bacteria Vibrio cholerae, using an extension of a traditional SIR compartment model; life history models of bacterial ageing; and individual based models (IBMs) of bacterial aggregation on a surface during the initial stages of biofilm formation.
